Ubuntu 16.04LTS + CUDA8.0 + Caffe2

本文主要是介绍Ubuntu 16.04LTS + CUDA8.0 + Caffe2，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

简介

caffe2：A New Lightweight, Modular, and Scalable Deep Learning Framework

caffe是一个清晰，可读性高，快速的深度学习框架。作者是贾扬清，加州大学伯克利的ph.D，现就职于Facebook。caffe的官网是http://caffe.berkeleyvision.org/。
目前，贾扬清和他在Facebook的团队正在开发新一代框架Caffe2。今年4月18日，Facebook开源了Caffe2。Caffe2与Caffe的区别是什么？Caffe2更注重模块化，在移动端、大规模部署上表现卓越。如同TensorFlow，Caffe2使用C++Eigen库，支持ARM架构。
用一个实用脚本，Caffe上的模型可轻易地被转变到Caffe2上。Caffe设计的选择使得它处理视觉类型的难题时很完美。Caffe2延续了它对视觉类问题的支持，且增加了对自然语言处理、手写识别、时序预测有帮助的RNN和LSTM支持。

caffe2宣称 CODE ONCE, RUN ANYWHERE

安装

环境

• Ubuntu 16.04 LTS
• CUDA8.0

参考资料

• 官方安装指南：caffe2 Install

注： 如今caffe2的安装将变得极为简单，通过Ubuntu的包管理器安装即可，可以直接下载 Pre-Built Binaries ，安装即可，这里不作说明，这里展示的是源码安装过程。

准备

参考官网说明，选择平台（Platform）和安装类型（Install type），可以得到具体安装说明，本文选择：

• Platform ： Ubuntu
• Install type ： Build From Source

注：如果使用非Python原生环境，如使用anaconda，请参考：anaconda install

基础环境配置

• 系统安装，可移步本人博客：U盘系统盘制作与系统安装（详细图解）
• CUDA安装，可移步本人博客：Ubuntu 16.04 LTS + CUDA8.0 + cudnn6.0 （新），或 NVIDIA DIGITS 学习笔记（NVIDIA DIGITS-2.0 + Ubuntu 14.04 + CUDA 7.0 + cuDNN 7.0 + Caffe 0.13.0）（旧）

依赖库安装

sudo apt-get update
sudo apt-get install -y --no-install-recommends \build-essential \cmake \git \libgoogle-glog-dev \libgtest-dev \libiomp-dev \libleveldb-dev \liblmdb-dev \libopencv-dev \libopenmpi-dev \libsnappy-dev \libprotobuf-dev \openmpi-bin \openmpi-doc \protobuf-compiler \python-dev \python-pip  # for Ubuntu 14.04
sudo apt-get install -y --no-install-recommends libgflags2
# for Ubuntu 16.04
sudo apt-get install -y --no-install-recommends libgflags-dev sudo pip install \future \numpy \protobuf

构建与安装

获取源码

从 这里 （或终端输入命令 git clone --recursive https://github.com/caffe2/caffe2.git && cd caffe2 ）下载 caffe2 源码并解压进入 caffe2 文件夹。

修改配置文件

caffe需要自己配置，对于caffe2，使用 cmake 配置，可 参考下面的构建安装部分。

添加环境变量

终端输入 sudo gedit ~/.bashrc 打开 “.bashrc” 文件，在文件末尾加入如下代码并保存：

#caffe2
#echo $PYTHONPATH
export PYTHONPATH=/usr/local:$PYTHONPATH
export PYTHONPATH=$PYTHONPATH:yourcaffe2rootdir/caffe2/build
#echo $LD_LIBRARY_PATH
export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH

然后重新打开一个终端，或者输入 source ~/.bashrc 加载新的环境变量。

构建安装

进入caffe2的根目录，终端执行如下命令，不报错误，代表安装成功！( -j$(nproc) 代表使用最大的线程编译，当然也可以手动指定，如 make -j4 )

# This will build Caffe2 in an isolated directory so that Caffe2 source is
# unaffected
mkdir build && cd build# This configures the build and finds which libraries it will include in the
# Caffe2 installation. The output of this command is very helpful in debugging
cmake ..# This actually builds and installs Caffe2 from makefiles generated from the
# above configuration step
sudo make install -j$(nproc)

执行完 cmake .. 终端输出：

$ echo $PYTHONPATH
/usr/local:yourcaffe2rootdir/caffe2/build
$ mkdir build && cd build
$ cmake ..
-- The CXX compiler identification is GNU 5.4.0
-- The C compiler identification is GNU 5.4.0
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Performing Test CAFFE2_LONG_IS_INT32_OR_64
-- Performing Test CAFFE2_LONG_IS_INT32_OR_64 - Success
-- Does not need to define long separately.
-- Performing Test CAFFE2_EXCEPTION_PTR_SUPPORTED
-- Performing Test CAFFE2_EXCEPTION_PTR_SUPPORTED - Success
-- std::exception_ptr is supported.
-- Performing Test CAFFE2_IS_NUMA_AVAILABLE
-- Performing Test CAFFE2_IS_NUMA_AVAILABLE - Success
-- NUMA is available
-- Performing Test CAFFE2_NEED_TO_TURN_OFF_DEPRECATION_WARNING
-- Performing Test CAFFE2_NEED_TO_TURN_OFF_DEPRECATION_WARNING - Success
-- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX2_EXTENSIONS
-- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX2_EXTENSIONS - Success
-- Current compiler supports avx2 extention. Will build perfkernels.
-- Build type not set - defaulting to Release
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Looking for pthread_create
-- Looking for pthread_create - not found
-- Looking for pthread_create in pthreads
-- Looking for pthread_create in pthreads - not found
-- Looking for pthread_create in pthread
-- Looking for pthread_create in pthread - found
-- Found Threads: TRUE  
-- Caffe2: Found protobuf with old-style protobuf targets.
-- Caffe2 protobuf include directory: 
-- Found Git: /usr/bin/git (found version "2.7.4") 
-- The BLAS backend of choice:Eigen
-- Could NOT find NNPACK (missing:  NNPACK_INCLUDE_DIR NNPACK_LIBRARY PTHREADPOOL_LIBRARY CPUINFO_LIBRARY) 
-- Brace yourself, we are building NNPACK
-- The ASM compiler identification is GNU
-- Found assembler: /usr/bin/cc
-- Found PythonInterp: /usr/bin/python (found version "2.7.12") 
-- Check if compiler accepts -pthread
-- Check if compiler accepts -pthread - yes
-- Caffe2: Cannot find gflags automatically. Using legacy find.
-- Found gflags: /usr/include  
-- Caffe2: Found gflags  (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libgflags.so)
-- Caffe2: Cannot find glog automatically. Using legacy find.
-- Found glog: /usr/include  
-- Caffe2: Found glog (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libglog.so)
-- git Version: v0.0.0
-- Version: 0.0.0
-- Performing Test HAVE_CXX_FLAG_STD_CXX11
-- Performing Test HAVE_CXX_FLAG_STD_CXX11 - Success
-- Performing Test HAVE_CXX_FLAG_WALL
-- Performing Test HAVE_CXX_FLAG_WALL - Success
-- Performing Test HAVE_CXX_FLAG_WEXTRA
-- Performing Test HAVE_CXX_FLAG_WEXTRA - Success
-- Performing Test HAVE_CXX_FLAG_WSHADOW
-- Performing Test HAVE_CXX_FLAG_WSHADOW - Success
-- Performing Test HAVE_CXX_FLAG_WERROR
-- Performing Test HAVE_CXX_FLAG_WERROR - Success
-- Performing Test HAVE_CXX_FLAG_PEDANTIC
-- Performing Test HAVE_CXX_FLAG_PEDANTIC - Success
-- Performing Test HAVE_CXX_FLAG_PEDANTIC_ERRORS
-- Performing Test HAVE_CXX_FLAG_PEDANTIC_ERRORS - Success
-- Performing Test HAVE_CXX_FLAG_WSHORTEN_64_TO_32
-- Performing Test HAVE_CXX_FLAG_WSHORTEN_64_TO_32 - Failed
-- Performing Test HAVE_CXX_FLAG_WFLOAT_EQUAL
-- Performing Test HAVE_CXX_FLAG_WFLOAT_EQUAL - Success
-- Performing Test HAVE_CXX_FLAG_FSTRICT_ALIASING
-- Performing Test HAVE_CXX_FLAG_FSTRICT_ALIASING - Success
-- Performing Test HAVE_CXX_FLAG_WZERO_AS_NULL_POINTER_CONSTANT
-- Performing Test HAVE_CXX_FLAG_WZERO_AS_NULL_POINTER_CONSTANT - Success
-- Performing Test HAVE_CXX_FLAG_WSTRICT_ALIASING
-- Performing Test HAVE_CXX_FLAG_WSTRICT_ALIASING - Success
-- Performing Test HAVE_CXX_FLAG_WD654
-- Performing Test HAVE_CXX_FLAG_WD654 - Failed
-- Performing Test HAVE_CXX_FLAG_WTHREAD_SAFETY
-- Performing Test HAVE_CXX_FLAG_WTHREAD_SAFETY - Failed
-- Performing Test HAVE_CXX_FLAG_COVERAGE
-- Performing Test HAVE_CXX_FLAG_COVERAGE - Success
-- Performing Test HAVE_STD_REGEX
-- Performing Test HAVE_STD_REGEX
-- Performing Test HAVE_STD_REGEX -- success
-- Performing Test HAVE_GNU_POSIX_REGEX
-- Performing Test HAVE_GNU_POSIX_REGEX
-- Performing Test HAVE_GNU_POSIX_REGEX -- failed to compile
-- Performing Test HAVE_POSIX_REGEX
-- Performing Test HAVE_POSIX_REGEX
-- Performing Test HAVE_POSIX_REGEX -- success
-- Performing Test HAVE_STEADY_CLOCK
-- Performing Test HAVE_STEADY_CLOCK
-- Performing Test HAVE_STEADY_CLOCK -- success
-- Found LMDB: /usr/include  
-- Found lmdb    (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/liblmdb.so)
-- Found LevelDB: /usr/include  
-- Found LevelDB (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libleveldb.so)
-- Found Snappy: /usr/include  
-- Found Snappy  (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libsnappy.so)
-- Found Numa: /usr/include  
-- Found Numa  (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libnuma.so)
-- OpenCV found (/usr/share/OpenCV)
CMake Warning at cmake/Dependencies.cmake:270 (find_package):By not providing "FindEigen3.cmake" in CMAKE_MODULE_PATH this project hasasked CMake to find a package configuration file provided by "Eigen3", butCMake did not find one.Could not find a package configuration file provided by "Eigen3" with anyof the following names:Eigen3Config.cmakeeigen3-config.cmakeAdd the installation prefix of "Eigen3" to CMAKE_PREFIX_PATH or set"Eigen3_DIR" to a directory containing one of the above files.  If "Eigen3"provides a separate development package or SDK, be sure it has beeninstalled.
Call Stack (most recent call first):CMakeLists.txt:101 (include)-- Did not find system Eigen. Using third party subdirectory.
-- Found PythonInterp: /usr/bin/python (found suitable version "2.7.12", minimum required is "2.7") 
-- Found PythonLibs: /usr/lib/x86_64-linux-gnu/libpython2.7.so (found suitable version "2.7.12", minimum required is "2.7") 
-- Found NumPy: /usr/local/lib/python2.7/dist-packages/numpy/core/include (found version "1.13.3") 
-- NumPy ver. 1.13.3 found (include: /usr/local/lib/python2.7/dist-packages/numpy/core/include)
-- Could NOT find pybind11 (missing:  pybind11_INCLUDE_DIR) 
-- Found MPI_C: /usr/lib/openmpi/lib/libmpi.so  
-- Found MPI_CXX: /usr/lib/openmpi/lib/libmpi_cxx.so;/usr/lib/openmpi/lib/libmpi.so  
-- MPI support found
-- MPI compile flags: 
-- MPI include path: /usr/lib/openmpi/include/openmpi/opal/mca/event/libevent2021/libevent/usr/lib/openmpi/include/openmpi/opal/mca/event/libevent2021/libevent/include/usr/lib/openmpi/include/usr/lib/openmpi/include/openmpi
-- MPI LINK flags path:  -Wl,-rpath  -Wl,/usr/lib/openmpi/lib  -Wl,--enable-new-dtags
-- MPI libraries: /usr/lib/openmpi/lib/libmpi_cxx.so/usr/lib/openmpi/lib/libmpi.so
CMake Warning at cmake/Dependencies.cmake:324 (message):OpenMPI found, but it is not built with CUDA support.
Call Stack (most recent call first):CMakeLists.txt:101 (include)-- Found CUDA: /usr/local/cuda-8.0 (found suitable version "8.0", minimum required is "7.0") 
-- Found CUDNN: /usr/local/cuda-8.0/include  
-- Caffe2: CUDA detected: 8.0
-- Found cuDNN: v6.0.21  (include: /usr/local/cuda-8.0/include, library: /usr/local/cuda-8.0/lib64/libcudnn.so)
-- Automatic GPU detection failed. Building for all known architectures.
-- Added CUDA NVCC flags for: sm_20 sm_21 sm_30 sm_35 sm_50 sm_52 sm_60 sm_61
-- Could NOT find NCCL (missing:  NCCL_INCLUDE_DIRS NCCL_LIBRARIES) 
-- Could NOT find CUB (missing:  CUB_INCLUDE_DIR) 
-- Could NOT find Gloo (missing:  Gloo_INCLUDE_DIR Gloo_LIBRARY) 
-- MPI include path: /usr/lib/openmpi/include/openmpi/opal/mca/event/libevent2021/libevent/usr/lib/openmpi/include/openmpi/opal/mca/event/libevent2021/libevent/include/usr/lib/openmpi/include/usr/lib/openmpi/include/openmpi
-- MPI libraries: /usr/lib/openmpi/lib/libmpi_cxx.so/usr/lib/openmpi/lib/libmpi.so
-- CUDA detected: 8.0
-- Found libcuda: /usr/local/cuda-8.0/lib64/stubs/libcuda.so
-- Found libnvrtc: /usr/local/cuda-8.0/lib64/libnvrtc.so
-- Found nccl: /home/liu/sfw/dlapp/caffe2/third_party/nccl/build/include  
CMake Warning at cmake/Dependencies.cmake:457 (message):mobile opengl is only used in android or ios builds.
Call Stack (most recent call first):CMakeLists.txt:101 (include)CMake Warning at cmake/Dependencies.cmake:533 (message):Metal is only used in ios builds.
Call Stack (most recent call first):CMakeLists.txt:101 (include)-- GCC 5.4.0: Adding gcc and gcc_s libs to link line
-- Include NCCL operators
-- Including image processing operators
-- Excluding video processing operators due to no opencv
-- Excluding mkl operators as we are not using mkl
-- Include Observer library
-- Using lib/python2.7/dist-packages as python relative installation path
-- Automatically generating missing __init__.py files.
-- 
-- ******** Summary ********
-- General:
--   CMake version         : 3.5.1
--   CMake command         : /usr/bin/cmake
--   Git version           : v0.8.1-1314-gd8770f8
--   System                : Linux
--   C++ compiler          : /usr/bin/c++
--   C++ compiler version  : 5.4.0
--   Protobuf compiler     : /usr/bin/protoc
--   Protobuf include path : /usr/include
--   Protobuf libraries    : optimized;/usr/lib/x86_64-linux-gnu/libprotobuf.so;debug;/usr/lib/x86_64-linux-gnu/libprotobuf.so;-lpthread
--   BLAS                  : Eigen
--   CXX flags             :  -O2 -fPIC -Wno-narrowing -Wno-invalid-partial-specialization
--   Build type            : Release
--   Compile definitions   : 
-- 
--   BUILD_BINARY          : ON
--   BUILD_DOCS            : OFF
--   BUILD_PYTHON          : ON
--     Python version      : 2.7.12
--     Python includes     : /usr/include/python2.7
--   BUILD_SHARED_LIBS     : ON
--   BUILD_TEST            : ON
--   USE_ATEN              : OFF
--   USE_ASAN              : OFF
--   USE_CUDA              : ON
--     CUDA version        : 8.0
--     CuDNN version       : 6.0.21
--     CUDA root directory : /usr/local/cuda-8.0
--     CUDA library        : /usr/local/cuda-8.0/lib64/stubs/libcuda.so
--     CUDA NVRTC library  : /usr/local/cuda-8.0/lib64/libnvrtc.so
--     CUDA runtime library: /usr/local/cuda-8.0/lib64/libcudart.so
--     CUDA include path   : /usr/local/cuda-8.0/include
--     NVCC executable     : /usr/local/cuda-8.0/bin/nvcc
--     CUDA host compiler  : /usr/bin/cc
--   USE_EIGEN_FOR_BLAS    : 1
--   USE_FFMPEG            : OFF
--   USE_GFLAGS            : ON
--   USE_GLOG              : ON
--   USE_GLOO              : ON
--   USE_LEVELDB           : ON
--     LevelDB version     : 1.18
--     Snappy version      : 1.1.3
--   USE_LITE_PROTO        : OFF
--   USE_LMDB              : ON
--     LMDB version        : 0.9.17
--   USE_METAL             : OFF
--   USE_MKL               : 
--   USE_MOBILE_OPENGL     : OFF
--   USE_MPI               : ON
--   USE_NCCL              : ON
--   USE_NERVANA_GPU       : OFF
--   USE_NNPACK            : ON
--   USE_OBSERVERS         : ON
--   USE_OPENCV            : ON
--     OpenCV version      : 2.4.9.1
--   USE_OPENMP            : OFF
--   USE_PROF              : OFF
--   USE_REDIS             : OFF
--   USE_ROCKSDB           : OFF
--   USE_THREADS           : ON
--   USE_ZMQ               : OFF
-- Configuring done
-- Generating done
-- Build files have been written to: yourcaffe2rootdir/caffe2/build

执行 make install -j$(nproc) 即开始编译构建，这个过程比较漫长，本人采用了30+线程，约10分钟。编译完出现如下提示信息：

[ 98%] Linking CXX executable ../bin/blob_test
CMakeFiles/blob_test.dir/core/blob_test.cc.o: In function `caffe2::(anonymous namespace)::ContentChunks_Serialization_Test::TestBody()':
blob_test.cc:(.text+0x3b3d6): warning: the use of `tmpnam' is dangerous, better use `mkstemp'
[ 98%] Built target blob_test
[ 98%] Linking CXX shared module python/caffe2_pybind11_state_gpu.so
[ 98%] Built target caffe2_pybind11_state_gpu
Scanning dependencies of target caffe2_detectron_ops_gpu
[ 98%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/sample_as_op.cc.o
[ 98%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/group_spatial_softmax_op.cc.o
[ 98%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/sigmoid_focal_loss_op.cc.o
[ 98%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/select_smooth_l1_loss_op.cc.o
[ 98%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/smooth_l1_loss_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/batch_permutation_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/sigmoid_cross_entropy_loss_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/spatial_narrow_as_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/roi_pool_f_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/upsample_nearest_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/ps_roi_pool_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/softmax_focal_loss_op.cc.o
[ 99%] Building CXX object modules/detectron/CMakeFiles/caffe2_detectron_ops_gpu.dir/affine_channel_op.cc.o
[100%] Linking CXX shared library ../../lib/libcaffe2_detectron_ops_gpu.so
[100%] Built target caffe2_detectron_ops_gpu
Install the project...
-- Install configuration: "Release"
CMake Error at cmake_install.cmake:36 (file):file cannot create directory: /usr/local/share/cmake/Caffe2.  Maybe needadministrative privileges.Makefile:104: recipe for target 'install' failed
make: *** [install] Error 1

可能会发现出现一个错误，是的，本人构建时执行的 make install -j$(nproc) ，因为没有权限，所以向 /usr/local/share/ 目录写入文件时报错，是的，执行 sudo make install -j$(nproc) 即可，

...-- Set runtime path of "/usr/local/bin/convert_db" to ""
-- Installing: /usr/local/bin/make_cifar_db
-- Set runtime path of "/usr/local/bin/make_cifar_db" to ""
-- Installing: /usr/local/bin/make_mnist_db
-- Set runtime path of "/usr/local/bin/make_mnist_db" to ""
-- Installing: /usr/local/bin/predictor_verifier
-- Set runtime path of "/usr/local/bin/predictor_verifier" to ""
-- Installing: /usr/local/bin/print_registered_core_operators
-- Set runtime path of "/usr/local/bin/print_registered_core_operators" to ""
-- Installing: /usr/local/bin/run_plan
-- Set runtime path of "/usr/local/bin/run_plan" to ""
-- Installing: /usr/local/bin/speed_benchmark
-- Set runtime path of "/usr/local/bin/speed_benchmark" to ""
-- Installing: /usr/local/bin/split_db
-- Set runtime path of "/usr/local/bin/split_db" to ""
-- Installing: /usr/local/bin/db_throughput
-- Set runtime path of "/usr/local/bin/db_throughput" to ""
-- Installing: /usr/local/bin/inspect_gpus
-- Set runtime path of "/usr/local/bin/inspect_gpus" to ""
-- Installing: /usr/local/bin/print_core_object_sizes
-- Set runtime path of "/usr/local/bin/print_core_object_sizes" to ""
-- Installing: /usr/local/bin/core_overhead_benchmark
-- Set runtime path of "/usr/local/bin/core_overhead_benchmark" to ""
-- Installing: /usr/local/bin/run_plan_mpi
-- Set runtime path of "/usr/local/bin/run_plan_mpi" to ""
-- Installing: /usr/local/bin/convert_encoded_to_raw_leveldb
-- Set runtime path of "/usr/local/bin/convert_encoded_to_raw_leveldb" to ""
-- Installing: /usr/local/bin/make_image_db
-- Set runtime path of "/usr/local/bin/make_image_db" to ""
-- Installing: /usr/local/bin/caffe2_benchmark
-- Set runtime path of "/usr/local/bin/caffe2_benchmark" to ""
-- Installing: /usr/local/bin/tutorial_blob
-- Set runtime path of "/usr/local/bin/tutorial_blob" to ""

caffe2默认被安装在根目录下/usr/local/，当然你也可以在配置时修改安装目录。

验证安装

参考 Test the Caffe2 Installation 验证安装。

为验证安装成功，终端输入 ：

cd ~ && python -c 'from caffe2.python import core' 2>/dev/null && echo "Success" || echo "Failure"

如果显示 Success 代表CPU版安装成功，如下图：

下面测试GPU，终端输入 python2 caffe2/python/operator_test/relu_op_test.py 却提示：

python2.7/dist-packages$ python2 caffe2/python/operator_test/relu_op_test.py
Traceback (most recent call last):File "caffe2/python/operator_test/relu_op_test.py", line 22, in <module>from hypothesis import given
ImportError: No module named hypothesis

缺少库，pip install hypothesis 即可：

perator Relu.
I0312 14:27:49.096366  5580 operator.cc:178] Engine CUDNN is not available for operator ReluGradient.
.
----------------------------------------------------------------------
Ran 1 test in 1.888sOK

输入如下命令，会显示显卡数量：

python2 -c 'from caffe2.python import workspace; print(workspace.NumCudaDevices())'

问题解决

出现错误，参考 Troubleshooting 解决。

安装

可视化

网络结构可视化

• 在线可视化： netscope
• tools/extra 下

训练可视化

在 tools/extra 下

权重可视化

参考 examples/notebooks/ 文件夹下文件，如：

• ‘net_surgery.ipynb’
• ‘00-classification.ipynb’

这篇关于Ubuntu 16.04LTS + CUDA8.0 + Caffe2的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！

---